*************************************************************************************************
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
*			                       README                                           *     		           	        	    
*       "The Effects of Government System Fractionalization on Satisfaction with Democracy"     *
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
*												*			
* Author:  	Pablo Christmann and Mariano Torcal            			                *
* Contact: 	pablo.christmann@upf.edu                           			        *
* Date: 	16th of May 2017                                  			        *                                           					
*												*
* * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * * *
*************************************************************************************************

Get started:

1. In order to run the do-files, you first need to create a replication folder, for example: "C:\PSRM Replication Files"
   You can change the directory of the replication folder by replacing "C:" and change the directory in the beginning of the "TSCS analysis.do"-file and the "CSES analysis.do"-file accordingly.


2. Copy and paste the replication material into the designated replication folder, for example into: "C:\PSRM Replication Files"


Description of the replication materials:

-  Our study rests on the analysis of two different datasets: A time-series cross-sectional (TSCS) panel dataset 
   and an individual level dataset based on pooled data from the Comparative Study of Electoral Systems (CSES).
   The replication material for the TSCS dataset is made of three files:
  	 #Dataset:  TSCS.dta
  	 #Syntax:   TSCS analysis.do
  	 #Log file: TSCS_analysis.log
   The replication material for the CSES dataset is made of three files:
  	 #Dataset:  CSES.dta
  	 #Syntax:   CSES analysis.do
  	 #Log file: CSES_analysis.log

-  There is no necessary order which do-file to run first. To follow the structure of the analysis presented in the article, it 
   is recommended to first run the file "TSCS analysis.do" and than the file "CSES analysis.do".

- "TSCS analysis.do" creates the following tables and figures: Table 1, Table D, Table F, Table H, Figure 1, Figure 2, Figure A, Figure B.
  "CSES analysis.do" creates: Table 2, Table E, Table G, Table I, Figure 3_Representation Deficit, Figure 3_Winning and Losing. 

-  The do-files require the following user written packages to be installed in Stata: outreg2, norm, labutil and collin. 

-  For all figures you can find the FORMATTED Stata graphs as .gph files as well as the orginial Stata output. Only the formatted graphs are presented in the article. 
   The unformatted and the formatted graphs carry the same information and only differ slightly in the presentation.

-  The out-files are created by outreg2 after the estimation of each model in Stata. 
   Notice: These files convey only part of the information necessary for the creation of the tables presented in the paper.
           The information for the variance components, Log likelihood, AIC and ICC have been taken directly from the Stata output and the post-estimation commands "estat ic" and "estat icc".

-  The do-files create various data sets with the suffix _temp. These data sets are of no substantive interest but necessary to run the .do files.